[Bug fortran/63205] [OOP] Wrongly rejects type = class (for identical declared type)

[dominiq at lps dot ens.fr]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63205

--- Comment #7 from Dominique d'Humieres <dominiq at lps dot ens.fr> ---
> Created attachment 34234 [details]
> A near final patch

First, the patch does not apply cleanly on trunk:

File to patch: gcc/fortran/trans-expr.c
patching file gcc/fortran/trans-expr.c
Hunk #1 succeeded at 3783 (offset 5 lines).
Hunk #2 succeeded at 3842 (offset 5 lines).
Hunk #3 succeeded at 3891 (offset 5 lines).
Hunk #4 succeeded at 3920 (offset 5 lines).
Hunk #5 succeeded at 3978 (offset 5 lines).
Hunk #6 succeeded at 4120 (offset 5 lines).
Hunk #7 succeeded at 4714 (offset 5 lines).
Hunk #8 FAILED at 5456.
Hunk #9 succeeded at 7391 (offset 5 lines).
Hunk #10 succeeded at 8267 (offset 5 lines).
Hunk #11 succeeded at 8341 (offset 5 lines).
Hunk #12 succeeded at 8387 (offset 5 lines).
1 out of 12 hunks FAILED -- saving rejects to file gcc/fortran/trans-expr.c.rej

I suspect the problem is spaces/tabs related in the line

        gfc_add_block_to_block (&se->post, &post);

After applying the hunk manually, I bootstrapped and regtested with the patch
for pr60255 without further trouble. The results are posted at

https://gcc.gnu.org/ml/gcc-testresults/2014-12/msg01222.html

and the failures for gfortran.dg/unlimited_polymorphic_1.f03 are not due this
patch, but the one for pr60255.

For the record, I have a patch in my working tree such that
gfortran.dg/widechar_intrinsics_10.f90 is miscomputed with -m32 (I did not yet
investigate the problem: too many patches in my working tree).

Thanks for working on this issue.